Inactivation of Salmonella, Listeria monocytogenes and Enterococcus faecium NRRL B-2354 in a selection of low moisture foods

    Inactivation of Salmonella, Listeria monocytogenes and Enterococcus faecium NRRL B-δ4 in a selection of low moisture foods Grzegorz Rachon, Walter Pe˜naloza, Paul A. Gibbs PII: DOI: Reference:

S0168-1605(16)30178-7 doi: 10.1016/j.ijfoodmicro.2016.04.022 FOOD 7206

To appear in:

International Journal of Food Microbiology

Received date: Revised date: Accepted date:

6 October 2015 12 April 2016 16 April 2016

Please cite this article as: Rachon, Grzegorz, Pe˜ naloza, Walter, Gibbs, Paul A., Inactivation of Salmonella, Listeria monocytogenes and Enterococcus faecium NRRL B-δ4 in a selection of low moisture foods, International Journal of Food Microbiology (2016), doi: 10.1016/j.ijfoodmicro.2016.04.022

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Grzegorz Rachona, Walter Peñalozab, Paul A. Gibbsa

RI P

T

Inactivation of Salmonella, Listeria monocytogenes and Enterococcus faecium NRRL B-2354 in a selection of low moisture foods Leatherhead Food Research, Randall’s Road, Leatherhead, Surrey KT22 7RY, United Kingdom

b

Nestlé Research Center, Lausanne, Vers-Chez-Les-Blanc, 1000 Lausanne 26, Switzerland

SC

a

NU

Abstract

The aims of this study were to obtain data on survival and heat resistance of cocktails of Salmonella, Listeria monocytogenes and the surrogate Enterococcus

MA

faecium (NRRL B-2354) in four low moisture foods (confectionery formulation, chicken meat powder, pet food and savoury seasoning) during storage before processing. Inoculated samples were stored at 16°C and cell viability examined at

ED

day 0, 3, 7 and 21. At each time point, the heat resistance at 80°C was determined. The purpose was to determine a suitable storage time of inoculated foods that can be

PT

applied in heat resistance studies or process validations with similar cell viability and heat resistance characteristics. The main inactivation study was carried out within 7

AC CE

days after inoculation, the heat resistance of each bacterial cocktail was evaluated in each low moisture food heated in thermal cells exposed to temperatures between 70 and 140°C. The Weibull model and the first order kinetics (D-value) were used to express inactivation data and calculate the heating time to achieve 5 log reduction at each temperature.

Results showed that the pathogens Salmonella and L. monocytogenes and the surrogate E. faecium NRRL B-2354, can survive well (maximum reduction < 0.8 log) in low moisture foods maintained at 16°C, as simulation of warehouse raw material storage in winter and before processing. The D80 value of the pathogens and surrogate did not significantly change during the 21 day storage (p>0.05). The inactivation kinetics of the pathogens and surrogate at temperatures between 70 and 140°C, were different between each organism and product. E. faecium NRRL B-2354 was a suitable Salmonella surrogate for three of the low moisture foods studied, but not for the sugar-containing confectionery formulation. Heating low moisture food in moisture-tight environments (thermal cells) to 111.2, 105.3 or 111.8°C can inactivate 5 log of Salmonella, L. monocytogenes or E. faecium NRRL B-2354 respectively.

1

ACCEPTED MANUSCRIPT 1. Introduction Although low moisture foods cannot support microbial growth, and were historically considered as ‘low risk’ in terms of pathogen contamination and no growth potential

T

compared to high water activity animal- or vegetal-derived products, they have

RI P

significantly contributed to the total number of food-borne infections and outbreaks (Chen, Freier, Kuehm et al., (GMA), 2009; Podolak et al., 2010; Beuchat, Mann & Alali, 2013). For example, it has been estimated that 1,000 people were infected by

SC

contaminated paprika present on potato chips in the 1993 outbreak in Germany (Lehmacher et al., 1995); over 400 cases (126 in 1981 and 283 in 2009) were associated with black pepper outbreaks (Gustavsen and Breen, 1984; Gieraltowski et

NU

al., 2013); and more than 200 cases were attributed to toasted oats cereal in the USA between April and June 1998 (Centers for Disease Control, 2001)

MA

Contaminated peanut butter was responsible for more than 400 cases in the USA between August 2006 and May 2007 (Centers for Disease Control, 2007), and again between 2008 and 2009 in 46 states resulting in a further 700 cases. It is generally

ED

recognised that many cases of food poisoning e.g. of salmonellosis, are unreported or not investigated, for all types of products; this in turn suggests that association of

PT

food-borne infections from dry ingredients, is much higher. According to RASFF (the Rapid Alert System for Food and Feed) a total of 477 notifications related to Salmonella in all types of food was recorded in 2014, of which 101 were related to

AC CE

low moisture foods – 21.2%. In 2015, 517 notifications were recorded of which 116 were related to low moisture food – 22.4%. Notifications related to Listeria are very much lower; 91 in 2014 with only two related to low moisture foods (butter and halva with pistachio nuts) and 99 in 2015 with three recorded notifications related to low moisture food (dry ham, sesame pasta and dried pork sausage). The high percentage of Salmonella notifications in low moisture foods indicates that current methods of harvesting e.g. of seeds, drying and primary processing for control or elimination of Salmonella, are not efficacious or are not correctly implemented. Attention should therefore be focused on improving harvesting methods, and evaluating the ability of pathogens to survive in low moisture foods both during storage and throughout processes. Appropriate and validated, processes and processing conditions should be developed and applied industrially, to control or eliminate food-borne pathogens in dry foods and ingredients for ready to eat products that are not heat treated before packaging and distribution. A number of studies relating to survival of pathogens in low moisture foods have been published, (e.g. Danyluk, et al., 2005; Uesegi, et al., 2006; Komitopoulou & Peñaloza, 2009;

2

ACCEPTED MANUSCRIPT Blessington, et al., 2012), but the product range, product composition, storage conditions and heating methods differ; therefore obtaining data for specific products, organisms and conditions is necessary. Although the number of cases of listeriosis is low, and those causally related to dry

T

foods, very low, the infection can be life-threatening (20-30% mortality). For this

RI P

reason, the survival and heat resistance of L. monocytogenes in a selection of four dried foods was investigated. This data is necessary for evaluating potential hazards and taking data-based decisions in HACCP studies. The use of clinical strains is a

SC

prudent option as there seems to be some evidence that strains isolated from foods and food-processing environments tend to exhibit reduced virulence (Liu et al.,

NU

2007).

While most publications show no limitations in using E. faecium NRRL B-2354 as a

MA

surrogate for Salmonella (Almond Board of California, 2007, 2014; Jeong et al., 2011; Enache et al., 2015) other studies have shown some limitations of using this surrogate (Rachon and Gibbs, 2015).

Survival curves obtained during heat inactivation studies are not always linear. Non-

ED

linear curves are very common in both laboratory experiments and in pilot plant scale trials, (Humpheson et al. 1998; Drosinos et al., 2006; Leguérinel et al. 2007). While

PT

for linear curves a first order kinetic model has been used and D- and z-values calculated, for non-linear curves, several different models have been proposed

AC CE

(Smith, 1991; Xiong et al. 1999; Juneja et al. 2001; Pasquali et al., 2016). The current study was undertaken to obtain data and information on the viability of two important pathogens, Salmonella and L. monocytogenes, in four dried food materials of different compositions during storage for 21 days (currently raw materials are generally ordered according to food processing schedules and kept temporarily in warehouses for short periods of time and therefore processed within 3 weeks), and to evaluate the utility of a non-pathogenic organism – E. faecium NRRL B-2354 – as a surrogate for these pathogens for potential use in food processing environments in case that biological validations/verifications are necessary to demonstrate that specific processing steps are appropriate to ensure food safety. Additionally, the heat resistance of the pathogens and surrogate in the four low moisture foods, was determined to evaluate the kinetics of inactivation to achieve a 5 log or greater inactivation levels.

2. Materials and Methods 2.1.

Low moisture foods

3

ACCEPTED MANUSCRIPT Four low moisture food formulations in powder form, were supplied by Nestec Ltd in sealed, flexible aluminised plastic pouches and stored at 16°C. The samples had been decontaminated at 25-50 kGy for this study at an external company. The composition of the products and water activity (aw) before inoculation are shown

T

in Table 1. In addition to the proximate composition, the confectionery formulation

RI P

contained starch (35%), sucrose (20%), maltodextrin (20%), wheat flour (20%), and natural flavouring ingredients (5%). The savoury seasoning contained salt (30%), glutamate (30%), sucrose (20%), rice flour, chicken meat, egg, spices. The chicken

SC

meat powder is an industrial raw material mix of chicken meat meal (85%) and salt (15%). The pet food formulation contained corn, rice, wheat flours (40%), and

NU

protein-rich materials like corn gluten, soybean meal, fish meal (35%), chicken byproduct meal (20%), mineral/vitamin premixes and natural flavouring (5%).

2.2.

Bacterial strains

MA

Table 1. Composition and aw of low moisture foods.

ED

A cocktail of six Salmonella strains was used in this study: S. Enteritidis PT 30; ATCC BAA-1045 (a strain associated with the first recorded food-borne outbreak

PT

linked to consumption of raw almonds, USA/Canada, 2001), S. Senftenberg 775W; ATCC 43845 (heat resistant in moist foods), S. Typhimurium; ATCC 14028 (chicken

AC CE

isolate), S. Anatum; ATCC BAA-1592 (a strain isolated from a tomatoes linked outbreak in the USA, 2004), S. Montevideo; ATCC BAA-710 (tomato isolate), S. Tennessee; K4643 (a human isolate from the 2006 peanut butter outbreak in the United States). These strains were selected for their survival above average among more than 30 strains in selected low moisture foods (data not shown). Selections was focused on the most frequently used strains with heat resistance above average, and strains from outbreaks linked to low moisture foods. All strains were obtained from American Type Culture Collection (ATCC) except for S. Tennessee K4643 which was supplied by Nestec Ltd. All strains were recovered on Tryptone Soya Agar (TSA, Oxoid CM0131) incubated aerobically for 18 h at 37 ± 0.5°C and a number of colonies (< 20) were dispersed in Cryo-preservation beads (TS/80-BL; TSC Ltd, Heywood, UK) containing Cryopreservative fluid: beef extract, peptone, sodium chloride, glycerol (20%), de-ionised Water. Three vials of each strain were prepared and stored at -70°C and used for preparing three independent replicates.

4

ACCEPTED MANUSCRIPT Preliminary screening of seventeen L. monocytogenes strains for ability to survive in low moisture foods identified five suitable strains with survival above average (data not shown). A cocktail of the five L. monocytogenes strains was used in this study: L. monocytogenes ATCC 15313 - 53 XXIII, DSMZ 20600 (serovar 1a, mammal isolate),

T

L. monocytogenes ATCC 49594 – Petite Scott A (serovar 4b human isolate, the

RI P

strain widely used as a reference strain for efficacy testing of food processing and preservation techniques, establishment of detection methods in foods, growth and heat resistance studies, and virulence studies, (Briers et

al., 2011), L.

SC

monocytogenes ATCC 35152 – NCTC 7973 (serovar 1a, isolated from mammal), L. monocytogenes ATCC 13932 - LMG 21264 (isolated from child with meningitis,

NU

Germany; serotype 4b), DSMZ 27575 (serovar 4b, human isolate) and L. monocytogenes - FRRB 2542 (Barotolerant salami isolate). Strains were obtained

MA

from ATCC and Leibniz-Institut DSMZ - Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH and L. monocytogenes FRRB 2542 was supplied by Nestec Ltd. All strains were grown and stored at -70°C as described above. A single strain of Enterococcus was used in this study: E. faecium NRRL B-2354

ED

(ATCC 8459) - (strain most frequently used in heat inactivation studies as a surrogate for Salmonella). This strain was obtained from ATCC and grown and

Inocula preparation

AC CE

2.3.

PT

stored at -70°C as described above.

This study was conducted using a cocktail of Salmonella strains, a cocktail of L. monocytogenes strains and an E. faecium NRRL B-2354 inoculum. The Salmonella cocktail combined all 6 strains (grown as individual cultures); L. monocytogenes cocktail combined all 5 strains (grown separately), and E. faecium NRRL B-2354 was used as a single strain inoculum. Previous studies have shown that cells prepared on lawns on agar plates are more resistant to heat than those prepared in broth (Uesugi et al. 2006; Komitopoulou and Peñaloza, 2009); the lawn plate technique described by Danyluk et al. (2005), Blessington et al. (2012) and the Almond Board of California (2014), was therefore adopted and used for preparation of both cocktails and the E. faecium NRRL B-2354 inoculum. Cocktails and E. faecium NRRL B-2354 inoculum were prepared as three independent trials. Strains were activated by inoculating 4 mL of brain heart infusion broth (BHI, Oxoid Ltd, Basingstoke UK) with 1 frozen bead followed by incubation at 37°C for 18-24 h. A second culture was prepared by inoculating 4 mL of BHI with 0.1 mL of the first culture and incubating at 37°C for 1824 h. The turbidity of each Brain Heart Infusion broth (BHI, Oxoid Ltd. Basingstoke, UK) culture after incubation provided a visual indication of adequate culture

5

ACCEPTED MANUSCRIPT activation. An aliquot (0.5 mL) of the 10-2 dilution of each of the second cultures was spread onto separate plates of Tryptone Soya Agar (TSA, Oxoid) and incubated aerobically at 37°C for 24 h. After incubation, cell lawns were harvested as a slurry by gently scraping the agar surface with a sterile L-shaped plastic spreader and 2 mL

T

of sterile Tryptone Salt diluent (TS). TS was prepared by mixing 1 g of Tryptone

RI P

powder (Oxoid) and 8.5 g of NaCl in 1 L of deionized water and autoclaved for 15 minutes at 121°C. Equal volumes from each of the cell slurries were mixed into a cocktail, vortexed for 10 seconds and used within 30 minutes. The slurries were

SC

mixed prior to the direct inoculation of the 4 low moisture foods. A dry inoculum techniques was not applied to avoid further dilution of initial viable counts. The initial

NU

source of low moisture food contamination incidents is mostly wet, and the equilibration or acclimatization time has been shown to be short in dry products

2.4.

Sample inoculation

MA

(Smith and Marks 2015).

Prior to inoculation, each of the low moisture food (LMF) samples were mixed by

ED

hand massage in large stomacher bags and 3 x 100 g replicate samples were evenly spread within large Petri dishes (140 mm diameter). Samples were placed in a safety

PT

cabinet and inoculated with 1 mL (1% v/w) of inoculum using needled syringes. Inoculation was conducted in two stages; first, 0.5 mL was distributed over the

AC CE

sample; the sample was carefully mixed and then the remaining 0.5 mL was added and mixed in the same way. Immediately after mixing, inoculated samples were sealed in stomacher bags and mixed by a vigorous external massage for 5 minutes in a safety cabinet until a lump free, homogenous mix was achieved. Inoculated samples were packed in aluminised plastic pouches and stored at 16°C. The inoculation method and mixing to achieve a homogenous distribution, was validated by enumeration of viable counts in at least triplicate sub-samples of inoculated powders in preliminary trials (SD <0.2 log CFU/g). Further confirmation of a homogenous distribution was obtained from enumeration of viable counts in food powders during storage (maximum SD <0.25 log CFU/g).

2.5.

Viability and changes in heat resistance during storage

The viability of Salmonella, and L. monocytogenes cocktails and E. faecium NRRL B2354 was evaluated during storage up to 21 days at 16°C following one day of moisture equilibration. At each time point, 1 ± 0.01 g of each replicate was weighed and serially diluted in TS diluent. Aliquots (0.1mL) of appropriate dilutions were spread on TSA plates. Plates were incubated aerobically at 37 ± 0.5°C and colonies

6

ACCEPTED MANUSCRIPT counted after 48 h. Periodically, colonies were confirmed by streaking on appropriate selective agars (Xylose lysine deoxycholate agar - for Salmonella and Oxoid Chromogenic Listeria Agar for Listeria) or using API or Microgen Listeria confirmation kits to confirm that no contamination had occurred during inoculations and laboratory

T

manipulations of the food powders.

RI P

In addition to evaluating viability during storage the heat tolerance at 80°C of the surviving population was evaluated. The temperature of 80°C was chosen after preliminary test showing inactivation levels in the middle range of 2-4 log reduction

SC

that would enable to quantitatively compare heat resistance changes between the selected storage times. At each time point, 1.2 ± 0.01 g of each replicate was

NU

weighed and placed into solid aluminium chambers (thermal cells) used for heat inactivation experiments. These thermal cells allowed to shorten come-up times

MA

when heating powders. Some of the thermal cells had incorporated built-in platinum thermocouples (Pt 100) designed and supplied by the Nestlé Research Center (Lausanne, Switzerland). Samples (1.2 g) were packed into the thermal cells (0.8 mm deep, 48 mm diameter). In each trial, temperature profiling was conducted and the

ED

core temperature of samples were recorded using a data logger (PicoLog TC-08; St Neots, UK).

PT

Temperature and time combinations used in this evaluation, were specific for the products and bacteria under investigation and it was expected that 2-4 log reductions

AC CE

would be achieved by the specific treatments. Changes of log reduction achieved at each time point would indicate if the bacteria heat resistance was changing or remaining stable during storage of the inoculated samples. After the heat treatment, the viable counts were carried out on TSA plates incubated aerobically at 37 ± 0.5°C for 48 h for Salmonella or 72 h for L. monocytogenes and E. faecium NRRL B-2354. 2.6.

Inactivation during low moisture food heat treatments at various temperatures

The heat resistance of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 was determined within the first week after inoculation following a minimum of 4 days of moisture equilibration at 16°C. Samples were weighed and placed into thermal cells and heat treated at four temperatures between 70 and 140°C depending on the product and organism under investigation. Heat inactivation experiments were performed using an oil bath (BAT3930; Grant Instruments, Cambridge, UK) filled with thermal conducting oil (Technical oil, VWR, Lutterworth, UK). In each trial a minimum of 5 log reductions was aimed for and achieved using at least five time points.

7

ACCEPTED MANUSCRIPT Experiments were conducted using three independent replicates. In addition each replicate was tested twice. Thermal cells containing inoculated material were submerged in the oil bath and held for the pre-selected times. Even though a circulating oil bath was used, preliminary trials indicated that times required to reach

T

target temperatures were significantly decreased when additional oil circulation was

RI P

introduced and so immersed thermal cells were moved back and forth at approximately 1 second intervals during the come-up time (approximately 1.5 - 2 minutes). Immediately after removing thermal cells from the oil, samples were cooled

SC

in cold water to temperatures below 30°C within 30 seconds. Following heat treatment, powders were removed from thermal cells and serial decimal dilutions

NU

prepared using Tryptone Salt diluent. Viable counts were enumerated within 10 minutes of preparing serial dilutions and volumes of 0.5 and 0.1 mL of appropriate

MA

dilutions were spread on the surface of TSA plates and incubated aerobically at 37 ± 0.5°C for 48 h for Salmonella and 72 h for L. monocytogenes and E. faecium NRRL B-2354. After each heat trial thermal cells and sealing rings were disinfected with Vircon (Day-Impex, Colchester, UK), washed twice using Greenline Plus - GP Mild

ED

Detergent and rinsed twice in sterile deionised water. Washed thermal cells were then dried in a drying cabinet at 55 ± 2°C for a minimum of 2 h. Preliminary trials

PT

confirmed that these treatments were effective in removing the inoculated organisms

AC CE

and no sterilisation (autoclaving) was necessary.

2.7.

Water activity and moisture content

Water activity (aw) and moisture content of inoculated samples was measured at the beginning of the trial (after one day of equilibration) and at the end of the storage at day 21. Water activities of the powders were measured using an AquaLab Series 3, aw meter (Decagon Devices, Inc. Pullman, USA) and two samples of each replicate were tested. Moisture content was measured using an Ohaus MB25 (Ohaus Corporation, Parsippany, USA) moisture tester at 133°C for 2 h. 2.8.

Data analysis

For each of the three independent storage trials, viable counts data of Salmonella, and L. monocytogenes cocktails and E. faecium NRRL B-2354 during storage in the four products, were expressed as mean log values (mean log) with standard deviations for each trial (SD). Changes (increase/decrease) in log values (Δ log) for each time point were compared to log values at day 0. The heat resistance in the low moisture food samples at 80°C of Salmonella, and L. monocytogenes cocktails and E. faecium NRRL B-2354 during the storage, was

8

ACCEPTED MANUSCRIPT expressed as level of inactivation (in mean log value) that was achieved at each sampling day (day 0, 3, 7 and 21) of the storage. The inactivation level was calculated by subtracting the mean log value of viable counts after the heat treatment from the mean log value of non-heat treated samples. In addition, D80 value (from

T

one time point data) at each storage time was calculated. Viable counts from

RI P

replicate heat inactivation trials were also expressed as mean log+/-SD and calculated for each time point.

SC

The heat inactivation curves of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in the low moisture food samples heated in thermal cells exposed to

NU

temperatures in the range of 70 - 140°C was used to calculate the heating time to reduce the initial population by 5 log. This time was calculated using Weibull model

MA

fitting (Boekel, 2002): log (N/N0) and expressed as a function of heating time (t) in the inactivation curves, where N = number of viable counts at time t, N0 = initial number of viable counts before heating. The following Weibull distribution equation was used

ED

to fit survival curves:

t - time (minutes)

PT

Where:

α - scale parameter (a characteristic time)

AC CE

β - shape parameter Weibull distribution corresponds to a concave upward survival curve if β <1, concave downward if β > 1 and reduces to an exponential (linear) distribution if β = 1. Parameters α and β were estimated using Excel equation solver and GRG (Generalized Reduced Gradient) Nonlinear Solving Method. Fitting of the model to raw data was confirmed by conducting an F-test using Excel (Microsoft Office) (Drosinos et al. 2006) and R2 (Brown, 2001). Parameters α and β were estimated for each replicate and mean values and standard deviations calculated. Time required to reduce the initial population of pathogens by 5 log (5D - as generally applied in the food industry) was calculated using the equation below (Van Boekel, 2002):

) Where: tD - time required to achieve required log reduction (minutes) d – number of required decimal log reductions (i.e. 5D = 5) α and β – parameters as described above

9

ACCEPTED MANUSCRIPT Time tD was calculated for each replicate separately and the average value (mean) and standard deviation (SD) was calculated. In addition, the standard log-linear model was fitted to the data of product at temperature higher than 70°C. The D values and time required to reduce the initial

T

population of pathogens by 5 log were calculated. Times calculated from Weibull

RI P

model and D values were then compared.

Microbial viability during storage of low moisture foods

NU

3.1.

SC

3. Results

MA

3.1.1. Water activity and moisture changes

During inoculation step, 1 % v/w of inoculum slurry was introduced to samples. This significantly changed aw of samples. The water activity of samples increased from

ED

0.434 to 0.565 for confectionery, 0.648 to 0.655 for seasoning, and 0.235 to 0.383 for chicken meat powder and from 0.576 to 0.653 for pet food. The moisture increased

PT

by an average of 0.82 % in all samples. The only small increase of water activity in seasoning can be explained by the fact that water introduced with inoculum quickly reacted with salt (salt concentration in seasoning was in a range of 20-30 %) and,

AC CE

therefore, the total available water (aw) content did not change significantly. During storage (21 days at 16°C), inoculated samples showed small but statistically significant differences (p<0.05) of water activity and moisture content. Maximum recorded changes of aw were; Δ aw = -0.040 and maximum changes of moisture were; Δ Moisture content = -0.61 %. Those changes were expected and no action was taken to stop them as it was believed that some changes may occur during storage and prevent them (storage in desiccators with adjusted humidity) would not be representative for ordinary storage.

3.1.2. Viability of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 during storage Salmonella, L. monocytogenes and E. faecium NRRL B-2354 survived within the same log level during the 21 day storage at 16°C in the inoculated low moisture foods (Figures 1A – 4A). The Salmonella viable counts were significantly lower

10

ACCEPTED MANUSCRIPT (p<0.01) at end of storage. However, the largest reductions observed were only 0.5 and 0.4 log in confectionery and pet food respectively. Normally, differences in viable cell counts of <0.5 log are generally regarded as non-significant in microbiological analysis (ISO 4833:2013). Salmonella viable counts at day 3 or 7 remained

T

statistically (p >0.05) at the same levels of inoculation.

RI P

The viable counts of L. monocytogenes decreased significantly (p<0.05) and progressively over storage and the largest reduction of 0.8 log observed was in the confectionery formulation by day 21. Reductions by day 3 and 7 in confectionery and

SC

pet food were similar and below 0.4 log. Surprisingly, in culinary seasoning viable counts remained stable over storage (p=0.317, p=0.580 and p=0.094 when

NU

comparing day 0 with respectively Day 3, Day 7 and Day 21). E. faecium NRRL B-2354 viable counts remained stable over storage (p>0.05) and

MA

its decrease was lower than 0.2 log amongst all products tested, indicating that this strain was a suitable fail-safe indicator for Salmonella or Listeria viability in low moisture foods upon storage before processing. Despite the fact that the culinary seasoning water activity (aw=0.655) after inoculation

ED

was the highest, Salmonella and L. monocytogenes survived notably better than in the other three products. In the culinary seasoning, only 0.2 log reduction was

PT

recorded for Salmonella and 0.1 log for L. monocytogenes. The survival of Salmonella and L. monocytogenes in culinary seasoning and chicken

AC CE

meat powder was slightly greater than in confectionery and pet food formulations during storage for 21 days at 16°C.

The results indicate that microbial viability during storage in dried foods, is dependent on the particular organism and can vary both with product composition and time. However the largest differences observed (Salmonella <0.54 log, L. monocytogenes <0.8 log), though statistically significant (p<0.01), have practically little impact on storage before processing of dry raw materials. Initial contamination levels are expected to remain at the same log level. The microbial viability within seven days of inoculation and storage at 16°C remained statistically (p>0.01) at the same level (decrease <0.4 log). Thus, short storage of maximum seven days of inoculated foods was adopted for the heat inactivation tests in this study.

3.1.3. D80 of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 during storage in low moisture foods.

11

ACCEPTED MANUSCRIPT Table 2. Changes in heat tolerance of Salmonella, Listeria and E. faecium in all products during storage.

T

Changes in the heat resistance of Salmonella, L. monocytogenes and E. faecium

RI P

NRRL B-2354 during storage in the four low moisture food samples, are shown in Table 2, where the p-values from statistical analysis are also summarized. The inactivation levels are expressed in log reduction (Δ log) achieved by heat treatment

SC

at 80°C of inoculated powders at one pre-selected time resulting in 2 to 4 log reductions.

This temperature of 80°C was selected from preliminary tests to inactivate only a

NU

fraction of the microbial population and obtain quantitative results to compare heat resistance over time after inoculation, and therefore determine the maximum keeping

MA

time at 16°C of inoculated samples throughout this study. The microbial inactivation in each product fluctuated randomly within a narrow range of variability (<0.5 log) between the different storage times (Table 2), and no significant correlation of D80

ED

value over time was observed, except for L. monocytogenes D80 increase (p <0.01) from 0.77 to 1.17 minutes in the confectionery formulation and a significant decrease

PT

(p<0.01) from 2.25 to 1.49 minutes in the culinary seasoning. These differences are relatively of minor relevance, and are similar to the standard deviation of the D80 values at other experimental conditions.

AC CE

The mean D80 values of Salmonella were greater in the chicken meat powder (8.3 ± 0.4 minutes) and confectionery formulation (6.8 ± 0.5 minutes) compared to culinary seasoning (1.8 ± 0.2 minutes) and pet food (0.71 ± 0.04 minutes). The D80 values of L. monocytogenes was greater in the culinary seasoning (2.06 ± 0.072 minutes) and chicken meat powder (2.0 ± 0.17 minutes) than in confectionery (0.94 ± 0.14 minutes) and pet food (0.62 ± 0.04 minutes). Salmonella had a greater heat resistance (D80 values) than L. monocytogenes in the low moisture foods tested except in the culinary seasoning where both D80 values are in the same range. The D80 values of the E. faecium NRRL B-2354 exceeded by approximately 3 to 4 times those of Salmonella in chicken meat powder, culinary seasoning and pet food. However, its D80 value in the confectionery formulation (4.62 ± 0.07 minutes) was lower than Salmonella (Table 2).

12

ACCEPTED MANUSCRIPT Both the viability and heat resistance of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 during 21 days of storage in all four products is shown in Figures 1 – 4 A, and B.

RI P

T

Fig. 1. Viability (1A) and heat resistance (1B) - (D80) of Salmonella, Listeria and E. faecium NRRL B-2354 in confectionery during storage.

NU

SC

Fig. 2. Viability (2A) and heat resistance (2B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in culinary seasoning during storage.

MA

Fig. 3. Viability (3A) and heat resistance (3B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in chicken meat powder during storage.

3.2.

ED

Fig. 4. Viability (4A) and Heat resistance (4B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in pet food during storage.

Inactivation during heat treatments at various temperatures

PT

The results showed that most of the microbial inactivation curves were log linear. However, a number of concave upward inactivation curves were observed and also a

AC CE

number of concave downward curves. Three examples of data fitting with the Weibull model to raw data are presented below (Figure 5): Knowing the parameters β and α, the heating time to achieve a 5 log reduction of each organism can be calculated from the following equation:

) Fig. 5. Examples of inactivation curves and fitting of Weibull Model. A; linear curve (Salmonella in seasoning at 80˚C), B; downward concave (Salmonella in seasoning at 120˚C), C; upward concave (Salmonella in confectionery at 100˚C). Replicate 1 (□), Replicate 2 (○), Replicate 3 (Δ) and (---) Weibull Model. Examples of inactivation curves and fitting of Weibull Model. A; linear curve, B; downward concave, C; upward concave

Both parameters α and β and their standard deviations values calculated for each set of survival curves from each product, bacteria and temperature, are shown in Table 3.

13

ACCEPTED MANUSCRIPT

RI P

T

Table 3. α and β parameters for Weibull model for Salmonella, L. monocytogenes and E. faecium NRRL B-2354 per product and temperature.

In some cases standard deviations were high (see Table 3) due to small variations between replicates that resulted in great differences in the parameters α and β. The

SC

greatest variation was observed for inactivation of L. monocytogenes in pet food at 80°C (Table 3). Heat inactivation curves between replicates were not significantly

NU

different (p>0.05). Some changes in the shape of survival curves were clearly visible (not shown) and thus significantly affected parameters α and β (Table 3). In this case additional adjustment of mean values of α and β were performed and calculated

MA

following the fitting of a Weibull model to mean values and not to individual replicates. Although adjusted parameters α and β significantly improved fitting of the Weibull model, standard deviations remained high. The Goodness of fit of the Weibull

ED

model to experimental data was confirmed by calculating R2 values and those are presented in Table 3. High values (R2 =0.931-1.000) indicated exceptionally good fit

PT

of the inactivation curves of minimum 5 data points, selected in preliminary tests for each temperature, product and organism. The heating time, including the come up time, to achieve a 5 log reduction in samples

AC CE

heated in thermal cells exposed to target temperatures was calculated using the Weibull model.

Applying the log-linear inactivation kinetics the traditional heat

resistance parameters D- and z- values were calculated from the inactivation curves. The early stages of these inactivation curves corresponded to increasing product temperature during the come up time and do not comply with sample isothermal conditions. In cases where R2 representing fitting of the linear inactivation was <0.95, linear regression curves were fitted by omitting the initial data point, that generally corresponded to sample temperature lower than 70°C. In addition, Table 4 shows the heating times for 5 log reductions in all products for all tested bacteria and set temperatures ranging from 70 to 120°C, where differences in heating times between organisms and products can clearly be observed. Table 4. Comparison of heating times (minutes) for a 5 log reduction calculated using Weibull model and first order kinetics (D-values).

14

ACCEPTED MANUSCRIPT Salmonella showed a significantly higher (p<0.5) heating time to reach a 5 log reduction than L. monocytogenes in high sugar formulation (confectionery) and high protein (chicken meat powder), whereas the heating time for Salmonella inactivation was in the same range as L. monocytogenes in the high salt-containing formulation

T

(seasoning) or a rich nutrient formulation (pet food) as shown in Table 4.

RI P

In a high sugar formulation (confectionery), the heating time for a 5 log reduction of Salmonella was significantly (p<0.05) longer (40.2 minutes at 80°C or 2 minutes at 100°C) than for E. faecium NRRL B-2354 (36.2 minutes at 80°C and 0.9 minutes

SC

100°C) (Table 4). However, this surrogate exhibited a significantly higher heating

NU

times than the pathogens in all other formulations

At low temperatures, e.g. 70 and 80°C, the differences in heating time for a 5 log

MA

reduction among the tested organisms was more noticeable, whereas at high temperatures such differences became smaller, indicating that the chemical composition or aw of products had more impact on heat resistance of bacteria than at higher temperatures.

ED

The p-values shown in Table 4 indicate differences between heating times to achieve 5 log reduction by using D values (first order kinetics) or Weibull Model. The heating

PT

time differences between the two models are statistically significant at temperatures above 100°C (p<0.01) except a few sporadic cases like E. faecium NRRL B-2354 in

AC CE

chicken meat powder.

3.2.1. Product temperature profile during heating Despite significant differences in chemical compositions of the low moisture foods tested in this study, the actual sample temperature logged during the thermal inactivation trials showed no significant differences in ramp up times and highest temperatures attained, amongst products. Temperature profiling conducted on all products at all temperatures, showed that samples were heated at the same rate and no significant differences were observed. At 1.5 minutes of heating time to 70 - 90°C, sample temperature was within 1°C of the target temperature and at 2 minutes of heating time, sample temperature is within approximately 0.5°C below the target temperature (Figure 6). At higher set temperatures (100, 120, 130 and 140°C) 5 log reduction was achieved during the come up time and before reaching the target temperature. For example; 5.5 log reduction of Salmonella was achieved in confectionery samples in thermal cells submerged in thermo-fluid bath at 120°C within 35 seconds, when product temperature was 111.2°C.

15

ACCEPTED MANUSCRIPT Fig. 6. Temperature of low moisture foods during heating in the thermal cells exposed to various set temperatures (70-140°C).

RI P

T

Table 5. Actual temperature of low moisture products to achieve a 5 log reduction of Salmonella, L. monocytogenes and E. faecium NRRL B-2354.

Product temperature recorded indicated that a 5 log inactivation occurred at different temperatures (and holding times at lower temperatures) depending on the product

SC

and bacteria. Table 5 shows the various heat treatment conditions to achieve a 5 log reduction in the four low moisture foods. Salmonella can be inactivated when heating pet food, seasoning, chicken powder and confectionery to 92.7, 96.0, 109.3 and

NU

111.2°C respectively (Table 5). L. monocytogenes was inactivated when heating pet food, confectionery, seasoning and chicken meat powder to 91.0, 96.6, 103.2 and

MA

105.3°C respectively (Table 5). E. faecium NRRL B-2354 was inactivated at slightly higher temperatures than Salmonella and L. monocytogenes except in confectionery; 104.3°C was required to inactivate E. faecium NRRL B-2354 and 111.2°C to

ED

inactivate Salmonella. The heating time to achieve a 5 log reduction of the three organisms at temperatures of 90°C and below is shown in Table 4.

PT

4. Discussion

This study has shown that Salmonella, L. monocytogenes and E. faecium NRRL B-

AC CE

2354 survived very well during for 21 days at 16°C in all four low moisture foods. This storage regime was a simulation of temporary storage in warehouses before processing; bacteria survive better in low moisture foods at low storage temperatures as documented by Komitopoulou and Peñaloza (2009), the counts of various Salmonella strains remained stable in cocoa butter oil at low temperatures. Rachon and Gibbs (2015) showed no significant reduction of Salmonella in paprika powder and rice flour during 21 days of storage at 16°C, and Uesugi et al.(2006), reported no decrease of Salmonella on almonds at low storage temperatures. Overall, survival of E. faecium NRRL B-2354 in low moisture foods confirmed that E. faecium NRRL B2354 was desiccation resistant and showed less reduction in viable counts than Salmonella and L. monocytogenes in low moisture foods during storage at 16°C for 21 days. The results in general, show that microbial viability during storage is dependent on the particular organism and can vary both with product composition and bacterial species.

16

ACCEPTED MANUSCRIPT Overall, the D80 values of Salmonella, L. monocytogenes and E. faecium NRRL B2354 in low moisture food samples did not change markedly during 21 days of storage. Small changes in heat resistance were observed for L. monocytogenes in seasoning and confectionery formulations. The heat resistance (D80) increase of L.

RI P

the decrease of 0.76 minutes in the culinary seasoning.

T

monocytogenes of 0.4 minutes in confectionery was statistically significant as well as

Inactivation curves obtained through the series of heat inactivation experiments confirmed that inactivation was not always linear. Non-linearity was greater at higher

SC

temperatures especially when microbial inactivation occurred during come up times, but non-linear inactivation curves also occurred at lower temperatures when come up

NU

time was not a significant fraction of the whole inactivation time. The Weibull model in these cases was shown to be an appropriate tool and times for a 5 log reduction can

MA

be calculated with precision.

Heat inactivation experiments indicated that there were some limitations of using E. faecium NRRL B-2354 as a surrogate, since in the sugar-containing confectionery formulation, heat resistance (D80) and time to reach a 5 log reduction of E. faecium

ED

NRRL B-2354 was shorter than for Salmonella at all tested temperatures. As demonstrated in many studies, the water activity of products has a significant

PT

impact on survival of bacteria during heat treatment; it was expected that survival of all tested bacteria during heating would be greatest in the inoculated chicken meat

AC CE

flour (aw=0.383) followed by confectionery (aw=0.565) and culinary seasoning (aw=0.655) or pet food (aw=0.653). While at lower heating temperatures (≤100°C) this general rule was confirmed in this study, at higher temperatures the 5 log reduction time for Salmonella was slightly greater in confectionery than in chicken flour, indicating that components of the confectionery formulation (sugars) may have a greater protective effect on Salmonella at higher temperatures. Protective functions of sugars is well documented; Sumner et al. (1991) showed that heat resistance of S. Typhimurium and L. monocytogenes increased as sucrose concentration increased and aw decreased; Mattick et al. (2001) also showed the great impact of sucrose and glucose-fructose solutions on heat resistance of Salmonella. Li et al. (2014) showed increased

heat

resistance

in

samples

which

contained

higher

levels

of

carbohydrates. They also observed that water activity was not the sole factor affecting the thermal resistance in those samples as the highest resistance of Salmonella was observed in samples with highest water activity, increased carbohydrate level and decreased fat concentration. Culinary seasoning, despite its high water activity (aw=0.655) was found to be the most protective product for L. monocytogenes. The 5 log reduction time at 80°C in culinary seasoning was the

17

ACCEPTED MANUSCRIPT highest when compared to other products including chicken meat powder. A 5-log reduction at a set heating temperature of 100°C required between ca. 1.0 and 3.5 minutes, for the three target organisms in the four dried food powders, considerably in excess of times and temperatures necessary for pasteurisation in high water

T

activity foods.

RI P

Salmonella showed a higher heat resistance than L. monocytogenes in the high sugar formulation (confectionery) and high protein (chicken meat powder), whereas the heat resistance of Salmonella was just slightly higher or not significantly different

SC

from L. monocytogenes in the high salt-containing formulation (seasoning) or the rich nutrient formulation (pet food).

NU

Comparison of heating times to achieve 5 log reductions calculated from the Weibull model and D-values showed significant differences (Table 4). At higher temperatures

MA

(≥100°C) calculation of heating time based on D-values, to achieve 5 log reduction were significantly lower than the times calculated using the Weibull model, because the initial heat shoulder until microbial inactivation was observed, is not taken into account, and the product has not yet reached the target temperature. This shows the

ED

inadequacy of forcing the application of first order kinetics when product temperature is increasing and when the holding times at target temperatures, cannot reliably be

PT

controlled, as in food processes like extrusion and continuous heat treatments

AC CE

without moisture evaporation.

The Weibull prediction was an appropriate mathematical model for fitting actual survival curves including the come up time and calculating more accurately 5 log reduction times than the traditional, forced linear kinetics based on D-values. Heating low moisture foods, similar to the ones used in this study, in moisture-tight environments (thermal cells) to 111.2, 105.3 or 111.8°C can inactivate 5 log of Salmonella, L. monocytogenes or E. faecium NRRL B-2354 respectively. Therefore using the Weibull model would be a more appropriate tool when inactivation kinetics of non-isothermal heating processes (e.g. extrusion) are assessed.

 Acknowledgements The project was funded by Nestlé Research Centre, Lausanne, Switzerland. The Research Centre also supplied the thermal cells used in heat resistance studies.

18

ACCEPTED MANUSCRIPT

 References

T

Almond Board of California, 2007. Guidelines for Validation of Dry Roasting Processes. http://www.almonds.com/sites/default/files/content/attachments/dryroast-validation-guidelines.pdf, (accessed 5th of September 2015)

SC

RI P

Almond Board of California, 2014. Guidelines for Using Enterococcus faecium NRRL B-2354 as a Surrogate Microorganism in Almond Process Validation. http://www.almonds.com/sites/default/files/content/attachments/guidelines_for_u sing_enterococcus_faecium_nrrl_b2354_as_a_surrogate_microorganism_in_almond_process_validation.pdf, (accessed 5th of September 2015)

NU

Beuchat, L.R., Mann, D.A & Alali, W.Q., 2013. Efficacy of sanitizers in reducing Salmonella on pecan nutmeats during cracking and shelling. Journal of Food Protection 76, 770-778.

MA

Blessington, T., Mitcham, E.J., Harris, L.J., 2012. Survival of Salmonella enterica, Escherichia coli O157:H7, and Listeria monocytogenes on Inoculated Walnut Kernels during Storage. Journal of Food Protection 75, 245–254. doi:10.4315/0362-028X.JFP-11-278

ED

Briers,Y., Klumpp, J., Schuppler, M.& Loessner, M.J. 2011. Genome sequence of Listeria monocytogenes Scott A, a clinical isolate from a food-borne listeriosis outbreak. Journal of Bacteriology 193, 4284–4285

PT

Brown, A.M., 2001. A step-by-step guide to non-linear regression analysis of experimental data using a Microsoft Excel spreadsheet. Computer Methods and Programs in Biomedicine 65, 191–200.

AC CE

Centers for Disease Control, 2007. Multistate Outbreak of Salmonella Tennessee Infections Linked to Peanut Butter (FINAL UPDATE). http://www.cdc.gov/salmonella/2007/peanut-butter-3-7-2007.html, (accessed 5th of September 2015) Centers for Disease Control, 2001. From the Centers for Disease Control and Prevention. Multistate outbreak of listeriosis--United States, 2000. JAMA: the Journal of the American Medical Association 285, 285–286. Chen, Y. et al., 2009. Control of Salmonella in Low-Moisture Foods. Pulse, 2009, 2333–2339. Danyluk, M.D., Uesugi, A.R., Harris, L.J., 2005. Survival of Salmonella enteritidis PT 30 on inoculated almonds after commercial fumigation with propylene oxide. Journal of Food Protection 68, 1613–1622. Drosinos, E.H., Mataragas, M., Vesković-Moracanin, S., Gasparik-Reichardt, J., Hadziosmanović, M., Alagić, D., 2006. Quantifying nonthermal inactivation of Listeria monocytogenes in European fermented sausages using bacteriocinogenic lactic acid bacteria or their bacteriocins: a case study for risk assessment. Journal of Foof Protection 69, 2648–2663. Enache, E., Kataoka, A., Black, D.G., Napier, C.D., Podolak, R., Hayman, M.M., 2015. Development of a Dry Inoculation Method for Thermal Challenge Studies in Low-Moisture Foods by Using Talc as a Carrier for Salmonella and a Surrogate Enterococcus faecium. Journal of Food Protection 78, 1106–1112. doi:10.4315/0362-028X.JFP-14-396 Gieraltowski, L., Julian, E., Pringle, J., Macdonald, K., Quilliam, D., Marsden-Haug,

19

ACCEPTED MANUSCRIPT

T

N., Saathoff-Huber, L., Von Stein, D., Kissler, B., Parish, M., Elder, D., HowardKing, V., Besser, J., Sodha, S., Loharikar, A., Dalton, S., Williams, I., Barton Behravesh, C., 2013. Nationwide outbreak of Salmonella Montevideo infections associated with contaminated imported black and red pepper: warehouse membership cards provide critical clues to identify the source. Epidemiology and Infection 141, 1244–52. doi:10.1017/S0950268812001859

RI P

Gustavsen, S., Breen, O., 1984. Investigation of an outbreak of Salmonella Oranienburg infections in Norway, caused by contaminated black pepper. American Journal of Epidemiology 119, 806–812.

SC

Humpheson, L., Adams, M.R., Anderson, W. a., Cole, M.B., 1998. Biphasic Thermal Inactivation Kinetics in Salmonella Enteritidis PT4. Applied and Environmental Microbiology 64, 459–464.

NU

Jeong, S., Marks, B.P., Ryser, E.T., 2011. Quantifying the performance of Pediococcus sp. (NRRL B-2354: Enterococcus faecium) as a nonpathogenic surrogate for Salmonella Enteritidis PT30 during moist-air convection heating of almonds. Journal of Food Protection 74, 603–609. doi:10.4315/0362-028X.JFP10-416

ED

MA

Juneja, V.K., Eblen, B.S., Marks, H.M., 2001. Modeling non-linear survival curves to calculate thermal inactivation of Salmonella in poultry of different fat levels. International Journal of Food Microbiology 70, 37–51. doi:10.1016/S01681605(01)00518-9

PT

Komitopoulou, E., Peñaloza, W., 2009. Fate of Salmonella in dry confectionery raw materials. Journal of Applied Microbiology 106, 1892–1900. doi:10.1111/j.13652672.2009.04144.x

AC CE

Leguérinel, I., Spegagne, I., Couvert, O., Coroller, L., Mafart, P., 2007. Quantifying the effects of heating temperature, and combined effects of heating medium pH and recovery medium pH on the heat resistance of Salmonella Typhimurium. International Journal of Food Microbiology 116, 88–95. doi:10.1016/j.ijfoodmicro.2006.12.016 Lehmacher, a, Bockemühl, J., Aleksic, S., 1995. Nationwide outbreak of human salmonellosis in Germany due to contaminated paprika and paprika-powdered potato chips. Epidemiology and Infection 115, 501–511. doi:10.1017/S0950268800058660 Li, C., Huang, L., Chen, J., 2014. Comparative study of thermal inactivation kinetics of Salmonella spp. in peanut butter and peanut butter spread. Food Control 45, 143–149. doi:10.1016/j.foodcont.2014.04.028 Liu, D., M. L. Lawrence, A. J. Ainsworth and F. W. Austin 2007. Toward an improved laboratory definition of Listeria monocytogenes virulence. International Journal of Food Microbiology 118, 101-115. Mattick, K.L., Jørgensen, F., Wang, P., Pound, J., Ward, L.R., Legan, J.D., Humphrey, T.J., 2001. Effect of Challenge Temperature and Solute Type on Heat Tolerance of Salmonella Serovars at Low Water Activity Effect of Challenge Temperature and Solute Type on Heat Tolerance of Salmonella Serovars at Low Water Activity. Applied and Environmental Microbiology 67, 4128–4136. doi:10.1128/AEM.67.9.4128 Pasquali, F., Klein, G., Reich, F., Manfreda, G., Valero, A., 2016. Modelling survival behaviour of Salmonella enterica ser. Enteritidis, Typhimurium and Tennessee

20

ACCEPTED MANUSCRIPT on table eggs during storage at different temperatures. Food Control 59, 314– 319. doi:10.1016/j.foodcont.2015.05.037

T

Podolak, R., Enache, E., Stone, W., Black, D.G., Elliot, P.H., 2010. Sources and Risk Factors for Contamination, Survival, Persistence, and Heat Resistance of Salmonella in Low-Moisture Foods. Journal of Food Protection 73, 1919–1936.

RI P

Rachon, G., Gibbs, P., 2015. Pathogens in low moisture food. Food Science and Technology 29, 45–48.

SC

Smith, R.L., 1991. Weibull regression models for reliability data. Reliability Engineering & System Safety 34, 55–76. doi:10.1016/0951-8320(91)90099-S

NU

Smith, D. F., Marks, B.P., 2015. Effect of rapid product desiccation or hydration on thermal resistance of Salmonella enterica serovar Enteritidis PT 30 in wheat flour. Journal of Food Protection 78, 281-286.

MA

Sumner, S.S., Sandros, T.M., Harmon, M.C., 1991. Heat Resistance of Salmonella Typhimurium and Listeria monocytogenes in Sucrose Solutions of Various Water Activities. Journal of Food Science 56, 1741–1743. Uesugi, A.R., Danyluk, M.D., Harris, L.J., 2006. Survival of Salmonella enteritidis phage type 30 on inoculated almonds stored at -20, 4, 23, and 35 degrees C. Journal of Food Protection 69, 1851–1857.

ED

Van Boekel, M. A. J.S., 2002. On the use of the Weibull model to describe thermal inactivation of microbial vegetative cells. International Journal of Food Microbiology 74, 139–159. doi:10.1016/S0168-1605(01)00742-5

AC CE

PT

Xiong, R., Xie, G., Edmondson, a. E., Sheard, M. a., 1999. A mathematical model for bacterial inactivation. International Journal of Food Microbiology 46, 45–55. doi:10.1016/S0168-1605(98)00172-X

Keywords:

Low moisture food Weibull model

E. faecium NRRL B-2354 Inactivation kinetics Pathogens Surrogate

21

ACCEPTED MANUSCRIPT

IP

T

Manuscript Tables and graphs

CR

Table 1. Composition and aw of low moisture foods. Confectionery

Seasoning

Chicken meat powder

Pet food

Moisture (%) aw Protein (N2% x 6.25) Fat (%) Carbohydrate (%)

8.36 0.434 3 1 87.5

8.95 0.648 24.2 1.2 26

3.63 0.235 69.5 25 3

10.94 0.576 30 6 53.8

AC

CE P

TE D

MA N

US

Composition

22

ACCEPTED MANUSCRIPT

Chicken meat powder aw=0.383

Pet food aw=0.653

IP

Listeria monocytogenes Δ log ± SD D80

0 3 7 21

80°C / 5 min -2.7 ± 0.1 -2.9 ± 0.3 -2.4 ± 0.1 -2.8 ± 0.1 p=0.015

1.85 1.71 2.05 1.77

0 3 7 21

80°C / 20 min -2.2 ± 0.1 -2.5 ± 0.1 -2.4 ± 0.0 -2.6 ± 0.1 p=0.039

0 3 7 21

80°C / 2 min -2.9 ± 0.2 -2.7 ± 0.1 -2.8 ± 0.1 -2.7 ± 0.1 p=0.148

80°C / 2 min -2.6 ± 0.2 -2.2 ± 0.3 -2.3 ± 0.1 -1.7 ± 0.1 p=0.006

E. faecium NRRL B2354 Δ log ± SD D80

0.77 0.91 0.89 1.17

80°C / 20 min -4.4 ± 0.1 -4.3 ± 0.1 -4.3 ± 0.1 -4.4 ± 0.0 p=0.179

4.59 4.70 4.68 4.52

80°C / 5 min -2.5 ± 0.1 -2.8 ± 0.1 -3.0 ± 0.0 -3.4 ± 0.1 p<0.001

2.25 1.80 1.66 1.49

80°C / 25 min -2.9 ± 0.1 -2.9 ± 0.1 -3.2 ± 0.0 -3.3 ± 0.1 p=0.126

8.66 8.52 7.78 7.63

8.93 8.03 8.30 7.79

80°C / 5 min -2.6 ± 0.1 -2.2 ± 0.1 -2.6 ± 0.1 -2.8 ± 0.2 p=0.029

1.95 2.31 1.92 1.82

80°C / 60 min -2.5 ± 0.2 23.75 -2.7 ± 0.2 22.62 -2.7 ± 0.2 21.88 -2.3 ± 0.1 25.71 p=0.120

0.67 0.74 0.71 0.75

80°C / 2 min -3.1 ± 0.3 -3.0 ± 0.5 -3.3 ± 0.1 -3.6 ± 0.2 p=0.211

0.64 0.66 0.62 0.56

80°C / 25min -2.9 ± 0.1 -2.4 ± 0.1 -2.6 ± 0.1 -2.8 ± 0.2 p=0.006

US

6.15 6.97 7.07 6.69

MA N

0 3 7 21

80°C / 20 min -3.3 ± 0.5 -2.9 ± 0.5 -2.8 ± 0.3 -3.0 ± 0.3 p=0.628

CR

D80

TE D

Culinary aw=0.655

Salmonella Δ log ± SD

CE P

Confectionery aw=0.565

Time (day)

AC

Product

T

Table 2. Changes in heat tolerance of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in all products during storage.

1.72 2.06 1.95 1.79

23

Salmonella L. monocytogenes

US

9.0

9.0

E. faecium

8.0

MA N

8.0

7.0

6.0 5.0

Salmonella

3.0

Listeria

2.0

E.faecium

0.0 0

5

10

15

Time (day)

CE P

1.0

TE D

4.0

20

25

D80 (min)

7.0

AC

Mean log (cfu/g)

CR

1B

1A

IP

T

ACCEPTED MANUSCRIPT

6.0 5.0 4.0 3.0 2.0 1.0 0.0 0

3

7

21

Time (day)

Fig. 1. Viability (1A) and heat resistance (1B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in confectionery during storage.

24

T

ACCEPTED MANUSCRIPT

IP

2B

Salmonella

9.0

12.0

8.0

US

CR

2A

L. monocytogenes E. faecium

MA N D80 (min)

6.0 5.0 4.0

Salmonella

TE D

3.0

8.0 6.0 4.0

L. monocytogenes

2.0

2.0

0.0 0

5

10

15

Time (day)

CE P

E. faecium

1.0

20

25

0.0 0

3

7

21

Time (day)

AC

Mean log (cfu/g)

10.0

7.0

Fig. 2. Viability (2A) and heat resistance (2B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in culinary seasoning during storage.

25

T

ACCEPTED MANUSCRIPT

IP

3B

Salmonella

CR

3A

30.0

E. faecium

US

9.0

L. monocytogenes

8.0

MA N D80 (min)

6.0 5.0 4.0

Salmonella

TE D

3.0

20.0 15.0 10.0

L. monocytogenes

2.0

0.0 0

5

10

15

Time (day)

CE P

E. faecium

1.0

20

25

5.0 0.0 0

3

7

21

Time (day)

AC

Mean log (cfu/g)

25.0

7.0

Fig. 3. Viability (3A) and heat resistance (3B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in chicken meat powder during storage.

26

IP

4B

9.0

2.5

8.0

US

CR

Salmonella L. monocytogenes E. faecium

2.0

MA N

7.0

D80 (min)

6.0 5.0 4.0

Salmonella

TE D

3.0

1.5

1.0

L. monocytogenes

2.0

0.5

1.0 0.0 0

5

10

15

Time (day)

CE P

E. faecium 20

25

0.0 0

3

7

21

Time (day)

AC

Mean log (cfu/g)

4A

T

ACCEPTED MANUSCRIPT

Fig. 4. Viability (4A) and Heat resistance (4B) - (D80) of Salmonella, L. monocytogenes and E. faecium NRRL B-2354 in pet food during storage.

27

IP

C 0.0

0.0 -1.0

-1.0

-3.0 -4.0 -5.0

-3.0 -4.0

US

-2.0

Log (N/N )

-5.0 -6.0 -7.0

-6.0

-8.0 0

2

4

6

8

10

12

0

MA N

Log (N/N )

-2.0

0.1

0.2

Time (min)

Time (min)

0.3

0.4

Log (N/N )

B

0.0

CR

A

T

ACCEPTED MANUSCRIPT

-1.0 -2.0 -3.0 -4.0 -5.0 -6.0 -7.0 0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

Time (min)

AC

CE P

TE D

Fig. 5. Examples of inactivation curves and fitting of Weibull Model. A; linear curve (Salmonella in seasoning at 80˚C), B; downward concave (Salmonella in seasoning at 120˚C), C; upward concave (Salmonella in Confectionery at 100˚C). Replicate 1 (□), Replicate 2 (○), Replicate 3 (Δ) and (---) Weibull Model. Examples of inactivation curves and fitting of Weibull Model. A; linear curve, B; downward concave, C; upward concave

28

ACCEPTED MANUSCRIPT

aw=0.565

Listeria monocytogenes β Mean ± SD

4.750 ± 1.233

1.136 ± 0.110

Mean ± SD

70 80

0.996

90

Pet food aw=0.653

0.109 ± 0.036

0.837 ± 0.070

0.998

120

0.309 ± 0.015

4.055 ± 0.281

0.990

140

0.260 ± 0.001

7.333 ± 0.299

0.989

80 90 100 120

0.763 ± 0.078 0.278 ± 0.015 0.310 ± 0.010 0.183 ± 0.006

0.970 ± 0.055 1.451 ± 0.064 3.299 ± 0.197 4.114 ± 0.145

1.000 0.999 0.996 0.995

1.378 ± 0.152

0.628 ± 0.013

0.449 ± 0.111 0.030 ± 0.008 0.221 ± 0.002 0.155 ± 0.019

70 80 90 100

80 90 100 120 130

β

E. faecium NRRL B-2354 R²

Mean ± SD

α

β

Mean ± SD

Mean ± SD

R²

0.106 ± 0.043

0.393 ± 0.017

0.086 ± 0.037

0.533 ± 0.031

0.997 0.993

0.280 ± 0.035

0.544 ± 0.011

0.981

0.173 ± 0.040

1.073 ± 0.099

0.998

0.142 ± 0.021

0.703 ± 0.027

0.995

0.140 ± 0.010

1.406 ± 0.047

0.999

0.246 ± 0.019

1.673 ± 0.076

0.993

0.211 ± 0.026

3.496 ± 0.499

0.958

0.687 ± 0.044 0.944 ± 0.035 2.198 ± 0.071 2.813 ± 0.068

0.997 0.992 0.986 0.996

7.272 ± 0.312 1.018 ± 0.086 0.507 ± 0.004 0.215 ± 0.017

1.545 ± 0.030 1.451 ± 0.054 3.069 ± 0.025 3.585 ± 0.327

0.996 0.991 0.993 0.936

0.983

0.442 ± 0.180

0.790 ± 0.085

0.991

4.498 ± 1.160

0.730 ± 0.053

0.980

0.696 ± 0.066 0.541 ± 0.025 2.853 ± 0.157 2.595 ± 0.257

0.995 0.997 0.993 0.979

0.195 ± 0.049 0.074 ± 0.025 0.232 ± 0.006

1.107 ± 0.113 1.146 ± 0.162 3.720 ± 0.152

0.995 0.996 0.977

0.816 ± 0.355 0.242 ± 0.071 0.251 ± 0.016

0.728 ± 0.096 0.908 ± 0.094 2.906 ± 0.138

0.993 0.999 0.990

0.189 ± 0.057 0.280 ± 0.026 0.225 ± 0.022

0.457 ± 0.035 0.904 ± 0.032 1.549 ± 0.080

0.984 0.947 0.992

0.157 ± 0.026 0.113 ± 0.132 0.185 ± 0.015

0.425 ± 0.009 0.620 ± 0.221 1.430 ± 0.071

0.998 0.931 0.990

2.680 ± 0.756 0.325 ± 0.036 0.109 ± 0.020

0.593 ± 0.039 0.631 ± 0.023 0.852 ± 0.060

0.998 1.000 0.998

0.266 ± 0.024

3.044 ± 0.268

0.998

0.182 ± 0.017

2.234 ± 0.173

0.994

0.270 ± 0.049

2.528 ± 0.378

0.978

TE D

0.338 ± 0.068 0.117 ± 0.015 0.214 ± 0.007 0.175 ± 0.005

CE P

Chicken meat powder aw=0.383

100

AC

Seasoning aw=0.655

α

R²

CR

α Mean ± SD

IP

Salmonella

US

Confectionery

Temp. (°C)

MA N

Product

T

Table 3. α and β parameters for Weibull model for Salmonella, L. monocytogenes and E. faecium NRRL B-2354 per product and temperature.

29

ACCEPTED MANUSCRIPT

Chicken meat powder aw=0.383

Pet food aw=0.653

0.768

1.99 ± 0.15 0.57 ± 0.01 0.36 ± 0.005

2.12 ± 0.09 0.25 ± 0.02 0.14 ± 0.02

0.249 0.000 0.000

80 90 100 120

9.51 ± 0.70 1.50 ± 0.05 0.65 ± 0.03 0.33 ± 0.004

9.55 ± 0.74 1.53 ± 0.07 0.27 ± 0.08 0.11 ± 0.003

0.956 0.579 0.001 0.000

80 90 100 120 130

67.27 ± 5.00 15.10 ± 1.38 2.73 ± 0.20 0.52 ± 0.02 0.44 ± 0.06

77.39 ± 3.64 17.76 ± 1.54 4.44 ± 0.16 0.24 ± 0.04 0.16 ± 0.01

70 80 90 100

39.53 ± 4.37 4.19 ± 0.34 1.09 ± 0.02 0.59 ± 0.01

56.55 ± 8.39 4.35 ± 0.28 0.89 ± 0.02 0.23 ± 0.01

CR

40.07 ± 1.83

US

40.57 ± 2.04

51.61 ± 13.58 7.99 ± 1.79 1.68 ± 0.05 0.79 ± 0.01

78.91 ± 7.60 11.23 ± 0.77 1.67 ± 0.05 0.62 ± 0.01

0.039 0.045 0.950 0.000

11.81 ± 0.94 1.55 ± 0.06 0.65 ± 0.003 0.42 ± 0.002

13.29 ± 0.82 1.53 ± 0.07 0.37 ± 0.01 0.11 ± 0.003

0.047 0.090 0.000 0.000 0.001

9.46 ± 0.69 1.76 ± 0.05 0.61 ± 0.04 0.45 ± 0.001

0.036 0.562 0.000 0.000

49.49 ± 8.46 4.62 ± 0.38 1.02 ± 0.02 0.54 ± 0.004

MA N

70 80 90 100 120 140

p - value

Listeria monocytogenes Weibull First order model kinetics Mean ± SD Mean ± SD p - value

TE D

Seasoning aw=0.655

Salmonella First order kinetics Mean ± SD

CE P

Confectionery aw=0.565

Weibull model Mean ± SD

Temp. (°C)

AC

Product

IP

T

Table 4. Comparison of heating times (mins) required to achieve 5 log reductions calculated using Weibull model and traditional first order kinetic approach (D-values). E. faecium NRRL B-2354 Weibull First order model kinetics Mean ± SD Mean ± SD p - value

24.95 ± 1.80 4.55 ± 0.13 1.06 ± 0.04 0.43 ± 0.01

36.17 ± 1.42 5.55 ± 0.13 0.89 ± 0.05 0.15 ± 0.01

0.001 0.001 0.009 0.000

0.110 0.757 0.000 0.000

35.34 ± 0.44 5.48 ± 0.13 1.12 ± 0.02 0.43 ± 0.01

28.90 ± 0.28 4.56 ± 0.08 0.46 ± 0.001 0.15 ± 0.004

0.000 0.001 0.000 0.000

10.34 ± 0.91 1.68 ± 0.10 0.59 ± 0.01 0.14 ± 0.02

0.158 0.280 0.397 0.000

126.2 ± 6.16 22.53 ± 2.05 3.52 ± 0.18 0.58 ± 0.01

133.5 ± 3.99 23.40 ± 1.53 3.65 ± 0.15 0.23 ± 0.02

0.158 0.591 0.397 0.000

15.53 ± 1.22 6.41 ± 1.25 0.80 ± 0.03 0.34 ± 0.01

0.002 0.079 0.000 0.000

162.2 ± 7.06 15.65 ± 0.56 1.91 ± 0.07 0.71 ± 0.02

190.4 ± 1.83 4.35 ± 0.28 2.04 ± 0.18 0.37 ± 0.01

0.003 0.000 0.327 0.000

30

CR

IP

T

ACCEPTED MANUSCRIPT

US

140

130

MA N

120

100

TE D

90 80

70 60

CE P

Temperature ( C)

110

50

30 20 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

AC

40

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2.0

Time (min) Fig. 6. Temperature of low moisture foods during heating in the thermal cells exposed to various set temperatures (70-140°C).

31

US

CR

IP

T

ACCEPTED MANUSCRIPT

Salmonella

L. monocytogenes

99.5 111.2

96.6

100 120

94.4 96.0

Chicken meat powder

100 120

99.7 109.3

Pet food

100

92.7

AC

Seasoning

TE D

Confectionery

Set temp. (°C) 100 120

CE P

Product

MA N

Table 5. Actual temperature of low moisture products recorded at the time taken to achieve a 5 log reduction in viable counts for Salmonella, L. monocytogenes and E. faecium NRRL B-2354 heated in thermal cells exposed to high set temperatures. E. faecium NRRL B-2354 98.5 104.3

94.4 103.2

98.8 104.3

93.2 105.3

99.8 111.8

91.0

95.4

32

ACCEPTED MANUSCRIPT

CE P

TE D

MA N

US

CR

IP

Pathogens or surrogate survived well in samples during storage (21 days at 16°C). Heat resistance did not change significantly throughout the storage period. Viability of pathogens or surrogate was adequate for inactivation/validation studies. E. faecium NRRL B2354 was a suitable surrogate in tested products except confectionery. Pathogens were inactivated by heating to 112°C solid foods in sealed thermal cells.

AC

    

T

Highlights:

33